The effect of soft skills, ethics, and value on the willingness of employers to continue recruiting UMT graduates

Article history: Received: August 10, 2020 Received in revised format: November 15 2020 Accepted: November 27, 2020 Available online: December 1, 2020 This study aims to investigate the effect of soft skills and ethics and value on the employers’ willingness to continue recruiting Universiti Malaysia Terengganu (UMT) graduates, together with the moderating effect of knowledge on the relationship between soft skills and the employers’ willingness to continue recruiting UMT graduates. The study’s respondents comprised of 208 employers in Malaysia who responded through an online survey using Google Forms. The survey data was then analyzed using the Partial Least Squares Structural Equation Modelling (PLS-SEM), indicating that soft skills positively affected the employers’ willingness to continue recruiting UMT graduates. Nevertheless, ethics and value were found to be insignificant factors on the employers’ willingness to continue recruiting UMT graduates. It was also revealed that knowledge had the moderating effect on the relationship between soft skills and the employers’ willingness to continue recruiting UMT graduates. Therefore, universities were recommended to invest in soft skills and knowledge education to ensure that graduates met the employers’ professional recruitment standards in areas of expertise. © 2021 by the authors; licensee Growing Science, Canada


Introduction
The transformation of knowledge, soft skills, and attitudes required in job recruitments and performances based on the modern economy (OECD, 2011) has extended the focus and emphasis on establishing specialised skills for UMT graduates in successful employment. One of the skills included in the curriculum is the development of soft skills to inculcate all the marketability criteria among UMT students, thus resulting in familiarisation with real-world challenges and allowing a head-start in the competitive market. Although thousands of degree holders graduate from various institutions, the procurement of a degree certificate did not guarantee employment. Moreover, the jobs offered by public and private sectors proved insufficient in fulfilling employment demands, inevitably increasing the unemployment rate of graduates from Malaysian higher educational institutions recently (Krish et al., 2012). Based on the Malaysian Department of Statistics, the number of unemployed graduates rose by 4.7%, from 154,900 in 2017 to 162,000 in 2018 (Department of Statistics, 2019), and should be duly acknowledged by higher educational institutions in developing talented graduates with the relevant skills and knowledge for the employment market (Mohamad Shukri et al., 2014). Geographically, UMT is one of the top public universities in the East Coast of Peninsular Malaysia specialising in marine sciences, aquatic resources, and oceanographic studies, in line with the university's slogan (Ocean of Discoveries, for Global Sustainability), to become Malaysia's leading and globally-respected marine institution. Additionally, UMT offers courses in maritime, economics, management, basic sciences, physical sciences, engineering, and computing. Furthermore, as the relative importance of educational skills and qualities may vary and evolve in a dynamic environment, graduates must be well-equipped and assertive to sustain a competitive edge. For example, public and private Malaysian universities producing graduates from similar fields should devise branding strategies among competing Malaysian graduates to secure employability and marketability in the job market (Mohd Abd Wahab & Ismail, 2014). Consequently, the UMT graduate marketability trend revealed an increase in percentage from 51.3% in 2015 to 87.4% in 2019 (CEC, 2020). In 2019, a total of 2,455 students secured employment upon graduation and were divided into four categories: working graduates, furthering studies, improving soft skills, and waiting for work placements. Out of the four categories, the number of employed graduates indicated the highest percentage of 79.9% (Laporan Kebolehpasaran Graduan UMT, 2019). Regardless, there was no guarantee concerning the willingness of current employees in recruiting UMT graduates. Hence, this research intended to analyse the extent to which factors, such as soft skills, knowledge, and ethics and value, determined the employers' willingness to continue recruiting UMT graduates. Specifically, the eight soft skills elements to be analysed include communication skill, leadership skill, technical skill, interpersonal skill, thinking skill, enterprise and entrepreneurial skill, teamwork skill, and information and technology skill.

Employer Willingness to Continue Recruiting Graduates
According to Lee and Kwon (2011), the term 'continuance intention' described a user's decision to continue using a specific technology previously used. For this study, intention referred to employers' willingness to continue recruiting university graduates. Therefore, employers with university graduates as the organisation's workforce expected professionalism in terms of good attitudes and ethics, soft skills, and knowledge in the area of expertise (Mohamad et al., 2020). Three factors were examined in this study to ascertain the positive influences of an employer's willingness to continue recruiting university graduates, namely, ethics and values, knowledge, and soft skills. In the event of a positive result through hypotheses testing, it could be concluded that the employers' willingness to continue recruiting university graduates were affected by the three factors.

Ethics and Values
Ethics is defined as the moral principles distinguishing right from wrong (Khan et al., 2013) and the individual's application of these moral values and codes in ethical dilemmas (Ndung' u, 2014). As ethical employers encouraged employees to be more committed to workplaces with strong ethical cultures, leadership, and environments (Loi et al., 2015), the employees' intentions to remain employed were likely to increase, with many studies confirming the positive relationship of values on intention or willingness behaviour. Additionally, as values reflected a positive relationship on the willingness to pay more for banking services (Izogo et al., 2020;Fernandez & Liu, 2019), it was deemed logical that being an ethical employee was one of the crucial factors in securing and sustaining a job. Therefore, employers who perceived employed individuals as ethical people with high moral standards may choose to continue recruiting from the same sources. Based on these findings, a positive effect on the employers' willingness to continue recruiting university graduates through ethics and values was expected. Thus, the following hypothesis was proposed: H1: Ethics and values have a positive effect on employers' willingness to continue recruiting UMT graduates.

Soft Skills
Soft skills referred to the necessary skills required for obtaining, retaining, and performing a job, with the potential of assisting employees in adjusting to various changes and increase the working capacity that complemented the needs of the working environment (De Guzman, 2013). Pitan (2017) reported that technological changes, rapid globalisation, and sectoral reforms demanded a change in the implementation of soft skills at the workplace. Individual job performances and career successes also depended on effective soft skills practices (Sangamitra & Priya, 2015). Therefore, graduates with good soft skills could view this criterion as a career investment following the high demand from employers (Robles, 2012). Regarding educational studies, soft skills were reported to have a positive effect on students' willingness to practice collaborative learning (Weinberger & Shonfeld, 2018;Demirtas & Akdogan, 2015). Therefore, the following hypothesis was proposed: H2: Soft skills have a positive effect on employers' willingness to continue recruiting UMT graduates.

Moderating Variable (Knowledge)
A moderator is an independent variable affecting the strength and direction of the relationship between another independent variable and a dependent variable. Also, the moderator interacts with the independent variable of interest to test the strength or weakness of the independent variable's relationship with the dependent variable at different levels of the moderator variable.
In other words, the relationship between the independent variable and the dependent variable relied on the moderator variable's value (Cohen & Cohen, 1983). As moderator variables were introduced in weak or inconsistent literature reviews, the establishment of knowledge as a moderator in the current study was essential. The previous study demonstrated that soft skills were positively related to the willingness to continue recruiting. Nonetheless, findings from Hinchliffe and Jolly (2011) revealed that soft skills were insignificant in employers' willingness to continue recruiting graduates. As such, the current study presented valid reasons to introduce a moderator. Knowledge referred to the acquisition of expertise and skills through a theoretical and practical understanding of certain issues (Che Ahmat et al., 2011;Ali et al., 2016;Shareef & Atan, 2019) and could be mastered by reading or practical exercises. According to Liebenberg et al. (2014), universities needed to provide assurance that knowledge was taught and developed for the job scopes suggested by industries, thus enabling future professionals to manage potential workplace challenges. Cai (2012) stated that university graduates required sound background knowledge in fulfilling various career demands and adjusting to the needs of the job market (AlMunifi & Aleryani, 2019). Hence, university graduates must possess practical and theoretical knowledge to successfully achieve organisational and business goals (Mills et al., 2010). Tasquier and Pongiglione (2017) also found that knowledge positively affected the willingness to act. As such, this study introduced knowledge as a moderator. Based on this finding, the following hypothesis was proposed: H3: The positive relationship between soft skills and employers' willingness to continue recruiting UMT graduates strengthens with a high level of knowledge.

Sampling Design and Data Collection
A quantitative approach was adopted to test the hypotheses of the research framework, with employers in Malaysia as the research respondents. A purposive sampling method was employed, as the study only focused on UMT graduates and concerned the theoretical effects of the research model (Hulland et al., 2017;Ngah, Ramayah, et al., 2019). The study data was gathered from employers at the Career and Entrepreneurship Centre, UMT. A total of 820 questionnaires were distributed online by email invitations using Google Forms, with 212 participants completing the survey. Out of the 212 questionnaires, four were invalid, leaving 208 valid surveys for analysis with a response rate of 25.4%. As proposed by (Hair et al., 2019;Ngah, Thurasamy, et al., 2019), the sample size of a study should be determined by analysis power, as the study employed PLS-SEM. At the power of 0.8, medium effect size, and P = 0.05, as suggested by (Gefen et al., 2011;Hafaz Ngah et al., 2020), the sample size would be calculated based on the model complexity. In accordance with (Green, 1991), a study's minimum sample size was 85. Hence, with 208 respondents for this study, the sample size complied with the minimum requirement. Furthermore, the questionnaire included the demographic characteristics of the organisation's and respondent's profile, and a soft skills scale: (enterprise and entrepreneurial skill, information computer technology (ICT) skill, leadership skill, technical skill, thinking skill, interpersonal skill, communication skill, teamwork skill), knowledge, ethics and value, and employers' willingness to continue recruiting graduates. Therefore, this study (Finch et al., 2012;Lievens & Sackett, 2012;Nickson et al., 2012) in measured soft skills as a higher-order construct and represented by leadership skill, ICT skill, entrepreneurship skill, technical skill, thinking skill, interpersonal skill, communication skill, and teamwork skill. The items of thinking skill, teamwork skill, ICT skill, and leadership skill were adopted from Parasuraman et al. (1985) and Ismail (2012), whereas communication skill and enterprise and entrepreneurial skills were adopted from Abdul Hamid et al. (2014). Technical skill was derived from Kavanagh & Drennan (2008), while interpersonal skill was adapted from Rasul et al. (2008). The items of ethics and value were adopted from Parasuraman et al. (1985), whereas items of knowledge were adopted from the Industrial Training Form, UMT (2013). Lastly, the items pertaining to the employers' willingness to continue in recruiting graduates were adapted from Venkatesh et al. (2012). The questionnaire in this study used a five-point Likert Scale from 1, representing "Strongly Disagree" to 5, representing "Strongly Agree" for each of the research model's construct. Fig. 2 illustrates the respondents' demographic profile. Based on the respondents, most of the organisations employed 21 to 50 staff (33.2%), 33.7% of the organisations employed five to 20 staff, 15.9% employed more than 100 staff, 8.7% employed less

Ethics & Value
Willingness to continue recruiting H1 H2 Soft Skills H3 Knowledge than five staff, 4.8% employed 50 to 70 staff, and 3.8% employed 71 to 100 staff. From the 208 respondents, 77.4% worked in private organisations, 15.9% were from government organisations, and 6.7% belonged to the 'others' category. The organisation's operational period in this study spanned more than 10 years (43.8%), five to 10 years (29.3%), and less than five years (11.1%), respectively.

Number of workers
Status of organization Years of operation Fig. 2. Respondent profile 4. Data Analysis

Common Method Bias (CMB)
The survey data from both the independent and dependent variables were derived from the same person at the same time (Ngah et al., 2018;Podsakoff et al., 2003). Procedural and statistical methods were applied to overcome the CMB issue of the study. For the procedural method, a different anchor scale was utilised to measure the independent and dependant variables (MacKenzie & Podsakoff, 2012). As for the statistical method, the study employed a Harman single factor test. Specifically, the first factor indicated 36.27%, which was below 40% as suggested by Fuller et al. (2016), hence confirming the low severity level in CMB for the study. Besides, a two-step approach by (Anderson & Gerbing, 1988) for the validation of the measurement and structural models would be applied. The establishment of the measurement model implied the concurrent establishment of convergent and discriminant validity. Therefore, the bootstrapping method, with 5,000 resampling techniques (Hair et al., 2019), would be implemented in the structural model phase. As the soft skills were measured to be a higher-order construct with type 1, (reflective-reflective mode), the repeated indicator approach would be applied for the higher-order construct.

Measurement Model
Convergent validity is a testing procedure applied to ensure that the multiple items used to measure the specific construct were genuine (Ngah et al., 2014;Abd Aziz et al., 2020;DeConinck, 2015). Thus, convergent validity would be confirmed if the loading and average variance extracted (AVE) was ≥ 0.5, and the composite reliability (CR) was ≥ 0.7 (Hair et al., 2019). Table 2 illustrates the results of the convergent validity measurements for the lower and higher-order study constructs. Based on all the loadings, AVEs and CRs were reported to be higher than the threshold value set up by Hair et al., (2019), thus indicating that all the research model constructs achieved the given requirements and confirmed the corroboration of convergent validity in the study. As such, Table 1 and Fig. 3 illustrates the convergent validity of the study results.   For discriminant validity, the Hetrotrait-Monotrait (HTMT) ratio was employed, as proposed by (Franke & Sarstedt, 2019;Murtaza et al., 2014). Therefore, discriminant validity would be confirmed if all the values read lower than 0.90, subsequently indicating the establishment of discriminant validity in this study. Table 2 illustrates the results of the HTMT ratio in the study. Prior to hypothesis testing, as proposed by (Hair et al., 2019), the study should confirm that multicollinearity did not limit the study, with the variance inflated factor (VIF) values to be ≤ 5 (Hair et al., 2017). As shown in Table 2, it was demonstrated that all the VIF values were lower than five, thus signalling that multicollinearity was not a serious issue in the study. Additionally, the bootstrapping technique produced the study results for the direct and moderating effects, as shown in Table 3 and Fig. 4, respectively. For the first hypothesis, ethics and values  employer willingness (β = -0.025, P = 0.426), the study found no relationship between ethics and value on the willingness of current employers to continue recruiting UMT graduates, hence supporting H1. This indicated the current employers' trust in UMT graduates' sense of ethics and value and did not utilise that factor anymore to continue recruiting UMT graduates. For H2, soft skills  employer willingness, (β = 0.307, P ≤ 0.05), the study found that soft skills positively affected the willingness to continue recruiting UMT graduates. In other words, high soft skills levels would enhance the willingness to continue recruiting UMT graduates. For the moderation effect, soft skills × knowledge  employer willingness, (β = 0.203, P ≤ 0.01), the study found that knowledge strengthened the relationship between soft skills and employer willingness, hence indicating that the combination of soft skills and good knowledge would increase the recruitment opportunities of UMT graduates by current employers. Hence, the H3 of the study was also supported. Fig. 5 demonstrates Dawson's Plot for the moderation analysis for a better perspective on the moderation effect.

Discussion and Conclusion
As employers' willingness to continue recruiting graduates from the same university is under-explored, this study attempted to discover the factors influencing employers' continuance behaviour to recruit UMT graduates through an online survey with the purposive sampling method. The study factors, such as ethics and values and soft skills, which were measured as a higherorder construct with eight dimensions, were tested. In order to enhance predictive power, this study introduced knowledge as a moderating factor between soft skills and the willingness to continue recruiting UMT graduates among current employers.
As such, two out of the three tested hypotheses were supported. For the H1, ethics and value were insignificant in the willingness to continue recruiting UMT graduates and demonstrated that the current employers already perceived UMT graduates to possess good ethics and values from previous recruitments, thus posing no issues to continue recruiting UMT graduates. Other studies by Ariffin et al. (2016) and Oyedele et al. (2018) also reported that value was insignificant in future behavioural intentions, thus supporting the study findings on ethics and value on the future perspectives of current employers. On the contrary, soft skills positively affected the willingness to continue recruiting UMT graduates. The finding revealed that graduates with good soft skills had a competitive advantage in career development and success, depending on the effectiveness of soft skills implementation (Robles, 2012;Sangamitra & Priya, 2015;McMurray et al., 2016). Therefore, UMT must enhance the graduates' soft skills to improve the chances of selection by potential employers during interview sessions. Although many efforts were undertaken concerning the development of soft skills, the industry players required the implementation of soft skills to match the industry's needs. Hence, initiating special sessions with expert industry players would boost the development of skills and confidence in demonstrating genuine potential for future employers. Sharing sessions with alumni also seemed beneficial, as alumni possessed a sound understanding of the students' culture, behaviours, values, and industrial requirements. Moreover, the third hypothesis confirmed that knowledge strengthened the positive effect of soft skills on the willingness to continue recruiting UMT graduates, indicating that a combination of good knowledge and soft skills would increase the chances of existing employers to continue recruiting UMT graduates. Graduates with good soft skills and knowledge on the area of study could improve recruitment opportunities provided by potential employers. Hence, the UMT management and lecturers must ensure the compatibility of the syllabi, notes, materials, and knowledge transferred to the graduates with industrial requirements.
The finding of this study would also enrich the university and graduates in understanding the importance of soft skills and knowledge in the recruitment process. Also, understanding the determinants of the graduates' continued recruitment would assist UMT to strategize on enhancing the graduates' marketability. More importantly, graduates must realise the value of soft skills and knowledge in succeeding without limitations. Therefore, universities should emphasise more on developing students' soft skills and knowledge in meeting the employers' requirements to obtain a job relevant to the area of study (Kee et al., 2012). Additionally, universities should focus more on training students to face the practicalities of employers' requirements, such as the relevant soft skills and strong basic knowledge, for graduates to obtain a job in the area of study. As this study was limited to the perspectives of UMT graduates', future research is recommended to conduct comparative studies between public and private university graduates in Malaysia.